329 - The Anti-Cyclic Citrullinated Antibody Titer Does Not Predict Disease Activity in Rheumatoid Arthritis

Nancy A. Shadick1, Roberta Glass1, Nancy E. Maher1, Michael E. Weinblatt1, Sonali DeSai1, Lori Chibnik1, Daniel H. Solomon1, Ronenn Roubenoff2, Alex Parker2, Elizabeth W. Karlson1. 1Brigham & Women's Hospital, Boston, MA; 2Millennium Pharmaceuticals, Cambridge, MA
Presentation Number: 329

Purpose: Autoantibodies to cyclic citrullinated peptides (anti-CCP) are associated with an increased risk of Rheumatoid Arthritis (RA) and may be a prognostic marker for erosive disease. However, little is known about whether quantitative CCP antibody titers are related to disease activity, making its use in clinical practice unclear. We evaluated whether CCP titer longitudinally predicts disease activity in RA, and whether this is modified by other clinical and genetic factors. Methods: Study subjects were RA patients enrolled in a long-term registry with yearly measures of demographics, functional status, disease activity, and C-reactive protein. Cases were genotyped by low resolution genotyping for the HLA-DRB1 shared epitope alleles (HLA-SE) and the putative protective alleles bearing the aspartic acid residue at position 70 (D70 alleles). General Estimated Equations (GEE) analysis of repeated measures assessed whether the anti-CCP titer (measured by the INOVA Quanta Lite CCP IgG ELISA) predicted measures of disease activity including DAS 28-CRP, multi-dimensional health assessment questionnaire (MDHAQ), RA disease activity index (RADAI), rheumatoid factor, physician global disease activity score and joint count scores. Results: In 525 RA patients (83% female, mean age 57.5 yrs, disease duration 15.9 yrs) the mean CCP titer was 124.5 (SD120). GEE analysis of the HLA-SE and D70 allele demonstrated that on average the anti-CCP titer was 28 points higher for each copy of the HLA-SE (p=0.0003) but not affected bu presence of the D70 allele. One year later, there was an average decline of 8 points in CCP titer for each copy of the D70 allele (p=0.005). GEE models adjusting for age, sex, disease duration, smoking, BMI, and medication use indicated that a change in anti-CCP titer over 1 year did not predict any of the 7 measures of disease activity. For example, the anti-CCP titer was not related to the baseline level of DAS28-CRP (beta=0.0005, p=0.21) nor was it related to the change in DAS28-CRP over 1 year (beta=0.0004, p=0.35); a 10 point increase in anti-CCP titer only translated to a 0.004 increase in the DAS-CRP. Further adjustment for the HLA-SE and the D70 allele did not modify the results. Conclusion: In this large clinical cohort, we found no relationship between the anti-CCP titer and traditional measures of disease activity even after adjusting for the effect of the HLA-SE and the D70 allele. The data suggest that the anti-CCP titer has little clinical utility as a marker of disease activity in RA.

  N.A. Shadick, Millennium Pharmaceuticals, 2; R. Glass, Millennium Pharmaceuticals, 2; N.E. Maher, millennium Pharmaceuticals, 2; M.E. Weinblatt, Millennium Pharmaceuticals, 2; S. DeSai, None; L. Chibnik, Millennium Pharmaceuticals, 2; D.H. Solomon, Millennium Pharmaceuticals, 2; R. Roubenoff, Millennium Pharmaceuticals, 3; A. Parker, Millennium Pharmaceuticals, 3; E.W. Karlson, None.